Amplification and amplification assays using DNA primers and a DNA polymerase are well-known for amplifying and for detecting nucleic acid target sequences. Methods for exponential amplification include the polymerase chain reaction (PCR), strand displacement amplification (SDA), nucleic acid sequence based amplification (NASBA), transcription-mediated amplification (TMA), and rolling circle amplification (RCA). Certain of these primer-dependent amplification methods, such as PCR, include thermal cycling, while others, such as NASBA, are isothermal. Among numerous DNA polymerases commonly used are Thermus aquaticus DNA polymerase (Taq polymerase) and reverse transcriptase. The design of linear DNA oligonucleotide amplification primers is generally accomplished with the aid of a computer program designed for that purpose. Among the available programs that can be utilized are PRIDE (Haas et al., Nucl. Acids Res. 26:3006-3012 1998); OLIGO (Rychlik et al., Nucl. Acids Res 17(21):8543-51 1989); OSP (Hilber et al., OSP: a computer program for choosing PCR and DNA sequencing primers. PCR Methods Appl. 1(2):124-128 1991); Primo (Li et al., Genomics 40(3):476-85 1997); and Primer Master (Proutski et al., Comput Appl Biosci 12(3):253-5 1996).
Nucleic acid amplification employing PCR is well known, as are assays that include PCR amplification. See U.S. Pat. Nos. 4,683,202, 4,683,195 and 4,965,188, and, generally, PCR PROTOCOLS, a guide to Methods and Applications, Innis et al. eds., Academic Press (San Diego, Calif. (USA) 1990). Homogeneous PCR assays that do not require washing to remove unbound detector reagents or probes and thus can be performed without opening amplification reaction vessels are also well known. Homogeneous PCR assays include both end-point assays, in which amplified product is detected at the end of the amplification reaction, and real-time assays, in which amplified product is detected during some or all of the thermal cycles as the reaction proceeds. See U.S. Pat. Nos. 5,994,056, 5,487,972, 5,925,517 and 6,150,097.
PCR amplification reactions, like other amplification methods referred to above, are generally designed to be symmetric, that is, to make double-stranded amplicons by utilizing a forward primer and a reverse primer that are “matched”; that is, they have melting temperatures that are as close as possible, and they are added to the reaction in equimolar concentrations. A technique that has found limited use for making single-stranded DNA directly in a PCR reaction is “asymmetric PCR.” Gyllensten and Erlich, “Generation of Single-Stranded DNA by the Polymerase Chain Reaction and Its Application to Direct Sequencing of the HLA-DQA Locus,” Proc. Natl. Acad. Sci. (USA) 85: 7652-7656 (1988); and U.S. Pat. No. 5,066,584. Asymmetric PCR differs from symmetric PCR in that one of the primers is added in limiting amount, typically 1-20 percent of the concentration of the other primer.
A more recently developed non-symmetric PCR amplification method is known as “Linear-After-The-Exponential” PCR or, for short, “LATE-PCR.” See Sanchez et al. (2004) PNAS 101: 1933-1938, Pierce et al. (2005) PNAS 102: 8609-8614, and published international patent application WO 03/054233 (3 Jul. 2003), which is incorporated herein by reference in its entirety. LATE-PCR takes into account the actual, concentration-adjusted melting temperatures of PCR primers at the start of amplification, referred to as Tm[0]. Tm[0] can be determined empirically, as is necessary when non-natural nucleotides are used, or calculated. A variety of fluorescent probes can be used with LATE-PCR, including, among others: molecular beacons, which are single-strands capable of forming a stem-loop structure that can close when not bound to target thereby bringing near to each other a fluorophore on one end and a quencher on the other end; linear single-stranded probes having a fluorophore on one end and a quencher on the other end; FRET probe pairs, which are two labeled, single-stranded probes that hybridize adjacently on a target sequence, permitting their labels to pass energy between them by FRET; fluorophore-labeled linear probes that FRET with a DNA dye; and linear double-stranded probes in which the fluorophore is on the strand that binds to target and the quencher is on a complementary strand that binds to the probe at an equivalent Tm in the absence of a target.
An undesirable feature of symmetric PCR amplifications is that, following the exponential phase of amplification, fluorescence curves obtained by monitoring replicate amplifications in real time diverge and plateau at different levels. Scatter indicates that replicates do not have the same reaction efficiency and reduces detection accuracy. This is a problem for PCR assays generally, but is particularly undesirable in the case of end-point assays. Scatter among replicates is considerably reduced but still present in LATE-PCR assays and asymmetric PCR assays, both of which have an exponential phase and a linear phase. The scatter in the linear phase in part reflects the scatter in the plateau at the end of the exponential amplification when the limiting primer runs out.
Another significant problem with primer-dependent amplification reactions, including PCR amplifications, is mispriming, which we consider to be manifested in several distinct types: Type 1, mispriming that occurs during preparation of reaction mixtures prior to the start of amplification; Type 2, mispriming that occurs during amplification if the temperature (which in PCR amplifications means the temperature in any thermal cycle) is for any reason reduced below the melting temperature of a primer; and Type 3, mispriming that occurs in the late stages of amplification, including a PCR amplification, that is continued after a high concentration of amplicon has been made. When Type 3 mispriming occurs in LATE-PCR and asymmetric reactions, the 3′ end of a single-stranded amplicon primes on another ss-DNA molecule, thereby converting ss-DNA into ds-DNA. Mispriming in a reaction can also result in scatter among replicate reactions. Mispriming includes primer-dimer formation, which can occur during any stage of amplification.
Several approaches have been used to address Type 1 mispriming. One approach is to modify the polymerase chemically so that it is inactive until heated to a high temperature such as 95° C. See U.S. Pat. Nos. 5,677,152 and 5,773,258. Another approach is to bind an antibody to the polymerase to inhibit the polymerase until the reaction is heated to a high temperature such as 95° C. to irreversibly denature the antibody. See U.S. Pat. No. 5,338,671. Chemically modified and antibody-bound DNA polymerases are commonly referred to as “hot start” DNA polymerases. Yet another “hot start” approach is to include an aptamer in the reaction mixture. See Doug and Jayasena (1996), J. Mol. Biol. 264: 268-278 and U.S. Pat. No. 6,020,130. An aptamer is a single-stranded oligonucleotide approximately 30 nucleotides in length that binds to a polymerase and inhibits its ability to extend a recessed 3′ end at low temperatures. Aptamers are not irreversibly denatured at 95° C., a typical highest temperature for a PCR cycle. Eppendorf-5 Prime, Inc. markets a proprietary ligand that is said to bind to Taq polymerase in a temperature-dependent manner and to inhibit its binding to double-stranded DNA at temperatures below about 50° C. Despite these many attempts, mispriming remains a problem with PCR amplifications.
Another type of mispriming during primer-dependent amplification reactions, including PCR amplifications, is known as primer-dimer formation and primer-dimer amplification. According to this phenomenon one primer hybridizes to the other primer or to another copy of itself and then undergoes extension of the 3′ end to generate a small double-stranded amplicon, which can then amplify further or can multimerize and amplify further. Primer-dimer formation can occur in the absence of target.
Quantitative analysis of amplification reactions, including PCR amplifications, has been enabled by real-time detection methods. In PCR amplifications the PCR cycle at which fluorescent signal becomes visible above the threshold cycle or CT of reactions is indicative of starting target concentrations. End-point analyses are semi-quantitative at best, due in part to scatter among replicates as the reaction exits exponential amplification. Electrophoretic analysis of double-stranded amplicons is semi-quantitative, and may utilize fluorescently labeled primers. End-point analysis utilizing fluorescently labeled probes, either allele-discriminating probes or mismatch-tolerant probes, are also semi-quantitative at best. By reducing scatter and producing single-stranded product, LATE-PCR offers significant improvement in end-point analysis, but scatter among replicates is often not completely eliminated, leaving quantitative and multiplex detection less accurate and more problematic than desired.
Design and construction of multiplex PCR assays often encounters the problem of mispriming, because the use of multiple pairs of primers in a single reaction geometrically increases the number of possible unintended interactions of primers and target sequences or other DNA strands that may be present. Indeed, in symmetric multiplex PCR assays it is very difficult to design all primer pairs to have the same melting temperature, and in a asymmetric or LATE-PCR multiplex PCR assay to design all of the limiting primers to have a single melting temperature and all of the excess primers to have a single melting temperature. It therefore follows that in a multiplex PCR assay the particular annealing temperature used for one or more thermal cycles is not likely to be optimal for all pairs of primers. If the primer annealing step of a PCR cycle is set to permit hybridization of the lowest Tm primer, the reaction will have reduced stringency for primers with Higher Tm's, which increases the chance for mispriming to occur. Moreover, in LATE-PCR assays the limiting primers used (whether in a monoplex or a multiplex) typically have melting temperatures 5° C. or more above the melting temperatures of the excess primers, again making it impossible to match a single primer annealing temperature to the melting temperature of both primers.
A property of DNA polymerases in primer-dependent amplifications, including PCR amplifications, is a nominal amount of selectivity, particularly a nominal ability to discriminate between a target sequence that is perfectly complementary to a primer and a sequence that is perfectly complementary except for a mismatch at the 3′ terminal nucleotide of the primer. It has been attempted to take advantage of this nominal selectivity to detect single-nucleotide mutations, or SNPs, by designing primers having their 3′ terminal nucleotide complementary to the target nucleotide that is subject to mutation. The amplification assay method known as the amplification refractory mutation system (ARMS) attempts to do that (Newton et al., Nucl. Acids Res. 17, 2503-2516 (1989); Wu et al., Proc. Natl. Acad. Sci. USA 86:2757-2760 (1989)). ARMS assays are prone to generation of false-positive signals due to mispriming and primer-dimer formation. Certain mispriming events may involve a primer that hybridizes incorrectly such that there is a 3′ mismatched nucleotide. Primer-dimer formation may also involve a mismatched 3′ nucleotide. In the last phase of a LATE-PCR amplification mispriming of a single-stranded amplicon on another single strand in the reaction mixture may also involve a mismatched 3′ nucleotide. Therefore, enhancing a polymerase's discrimination against a 3′ terminal mismatch can, among other effects, reduce mispriming. Attempts have been made to improve selectivity during amplification beyond the foregoing nominal selectivity by making amplification primers themselves more selective. For example, Tyagi et al. added to the 5′ end of a primer a sequence complementary to the 3′ end of the primer to form a stem-loop structure wherein the loop and the 3′ portion of the stem are complementary to the target strand (U.S. Pat. No. 6,277,607). This approach is not seen to reduce the difficulty, described above, of designing primers for multiples assays. Making primers more selective does not, of course, improve the selectivity of DNA polymerases.
To improve selectivity, an alternative to modifying primers is to affect the DNA polymerase itself. U.S. patent application Ser. No. 11/242,506 describes a class of reagent additives that somewhat improve product specificity and that greatly reduce or in some cases practically eliminate the effects of mispriming in PCR amplification reactions. This class of reagents is comprised of single oligonucleotides molecules that are able to fold into hairpin structures having a stem and a loop when the temperature is lowered below the melting temperature of the stem. Although the double-stranded stem closes, the nucleotides at the 3′ and 5′ ends tend to unwind. Therefore, these additive reagents are chemically modified at both their 3′ and 5′ ends to keep the ends closed. End closure in this way effectively increases the melting temperature of the stem. In the closed configuration these reagent additives interact with DNA polymerase so as to improve selectivity. In the closed configuration they also inhibit polymerase activity of DNA polymerases. While these additives out-perform existing “hot-start” methodologies in all types of PCR and can be used to prevent the accumulation of undesired products, including primer-dimers and misprimed amplicons, both at early stages of the reaction and during LATE-PCR reactions having many cycles (typically 60 cycles and more), they do have their limitations which are inherent to their being comprised of a single oligonucleotide. Specifically, the length of the stem cannot be greater than about 12 nucleotides, because, if it is, and is also chemically modified at its ends, its melting temperature becomes so high that it does not readily open when the PCR is heated to the extension temperature. Even when added at low concentration, hairpin molecules with long stems and high Tm tend to inhibit the reaction. Yet another difficulty inherent to these additives is that they are not linearly symmetric, i.e. one end of the closed hairpin is open while the other end is a loop. As described in U.S. patent application Ser. No. 11/242,506, molecules with loops comprised of 3-22 nucleotides tend to inhibit amplification more readily than molecules in which the loop is formed by use of a 3 carbon or 6 carbon linker. It would be desirable to have reagents which are structurally symmetrical end-to-end.
Kainz et al. (2000) Biotechniques 28: 278-282 reported that DNA fragments, double-stranded DNA oligonucleotides, having lengths of 16-21 nucleotides can inhibit mispriming that occurs at or just below the optimal annealing temperature of symmetric PCR reactions and thereby prevent amplification of non-specific products. The DNA oligomers are reversibly denatured during the melting step of the PCR cycling. In all cases the assays that Kainz et al. employed revealed the presence of, and inhibition of, mispriming that takes place when the temperature is descending to the optimal annealing temperature after the first melting event at 95° C. This does not address Type 1 mispriming, as Kainz et al. acknowledged, and their data reveal that double-stranded fragments that are only double-stranded when Type 1 mispriming occurs (that is, with melting temperatures >5° C. below the annealing temperature of the reaction) fail to prevent mispriming. From Kainz et al. one infers that their method will likely be even more unreliable in multiplex reactions because, as explained above, the annealing temperature cannot simultaneously be optimized for all pairs of primers. Kainz et al. also acknowledged that, although they did not observe it in their particular experiments, double-stranded DNA oligonucleotides may trigger mispriming, if they become the target for hybridization of one or more primers in the reaction.